Venkatesh S. Kadandale

वेङ्कटेश शेनोई कडन्दले

CV Available Upon Request

Hey, I'm Venkatesh!

I am a computer vision researcher, software developer and a musician based in Barcelona.

I received my Ph.D. at Universitat Pompeu Fabra (UPF) , Barcelona, under the supervision of Prof. Gloria Haro. My research primarily focusses on developing AI systems to perform intelligent tasks like source separation and synchronization by exploiting the rich audio-visual correspondence between the audio-visual modalities. Lately, I started working on voice cloning systems.

Interests

Voice Cloning
Audio-visual Source Separation
Audio-visual Synchronization
Deep Learning
Music Information Retrieval

Education

Ph.D. in Info. and Communication Tech., 2022

Universitat Pompeu Fabra
Masters in Sound and Music Computing, 2018

Universitat Pompeu Fabra
Bachelors in Electrical and Electronics Eng., 2015

National Institute of Technology Karnataka

News

[Feb 2023]

I have joined Quantic Brains as an AI Engineer.

[Sep 2022]

Successfully defended my PhD thesis titled Towards audio-visual deep learning methods for singing voice separation and lip synchronization with honor (Cum Laude).

[Jul 2022]

Paper accepted to ECCV 2022.

[Jun 2022]

Paper accepted to Interspeech 2022.

[May 2022]

Two papers accepted to Sight and Sound workshop, CVPR 2022.

[Apr 2022]

Participated in #HiloTesis for dissemination of my Ph.D. research. [Link]

[Dec 2021]

New Single Dinner at the Glaciers out! [Link]

[Oct 2021]

Paper accepted to BMVC 2021.

[Jun 2021]

Paper accepted to Sight and Sound workshop, CVPR 2021. [Video]

[Jul 2020]

Paper accepted to IEEE MMSP 2020 for Oral Presentation. [Talk]

[Jun 2018]

Awarded INPhINIT doctoral fellowship for pursuing Ph.D. in Spain.

Publications

Towards Audio-Visual Deep Learning Methods for Singing Voice Separation and Lip Synchronization

Venkatesh S. Kadandale

Ph.D. Thesis [September 2022], Universitat Pompeu Fabra, Barcelona

PDF Presentation

VoViT: Low Latency Graph-based Audio-Visual Voice Separation Transformer

Juan F. Montesinos, Venkatesh S. Kadandale, Gloria Haro,

Accepted to ECCV 2022

PDF Webpage Demos

VocaLiST: An Audio-Visual Synchronisation Model for Lips and Voices

Venkatesh S. Kadandale, Juan F. Montesinos, Gloria Haro,

Interspeech 2022

PDF Webpage Code Presentation Poster Demos

A cappella: Audio-visual Singing Voice Separation

Juan F. Montesinos^*, Venkatesh S. Kadandale^*, Gloria Haro,

British Machine Vision Conference (BMVC) 2021

PDF Dataset Webpage Demos

Multi-channel U-Net for Music Source Separation

Venkatesh S. Kadandale, Juan F. Montesinos, Gloria Haro, Emilia Gómez

MMSP 2020

PDF Code Webpage Presentation

Musical Instrument Recognition in Multi-Instrument Audio Contexts

Venkatesh S. Kadandale

Master Thesis [September 2018], Universitat Pompeu Fabra, Barcelona

(*) equal contribution

Presentations

Synchronisation of Lips and Voices

Venkatesh S. Kadandale, Juan F. Montesinos, Gloria Haro,

CVPR Workshop - Sight and Sound, 2022

PDF Webpage Demos

Audio-visual voice separation transformer

Juan F. Montesinos, Venkatesh S. Kadandale, Gloria Haro,

CVPR Workshop - Sight and Sound, 2022

PDF Webpage Demos

Estimating Individual A Cappella Voices in Music Videos with Singing Faces

Venkatesh S. Kadandale^*, Juan F. Montesinos^*, Gloria Haro,

CVPR Workshop - Sight and Sound, 2021

PDF Demos Presentation

(*) equal contribution

Teaching

[Jan - Mar 2022]

Calculus II, Universitat Pompeu Fabra, Barcelona.

[Jan - Mar 2022]

Linear Algebra, Universitat Pompeu Fabra, Barcelona.

[Jan - Mar 2020]

Calculus II, Universitat Pompeu Fabra, Barcelona.

[Jan - Mar 2019]

Calculus II, Universitat Pompeu Fabra, Barcelona.

Contact

venkatesh [dot] kadandale [at] gmail [dot] com
08003 Barcelona, Spain